GitHub Repository

Git-anchored decentralised intent(conversation) ledger for teams who build with AI

4 starsGo

Rekal – RAG-queryable intent ledger anchored to Git

Name: Rekal – RAG-queryable intent ledger anchored to Git
Availability: InStock
Author: guocongwudi

by guocongwudi·Feb 27, 2026·1 point·0 comments

Visit Project View on HN

AI Analysis

●●●BangerWizardryZero to OneBig Brain

Agents remember *why* code exists, all local git-native, no cloud lock-in.

Strengths

•True zero-to-one problem: AI agents have no institutional memory; Rekal solves via append-only intent ledger
•Hybrid search (BM25+LSA+embeddings) runs entirely local in single binary—no API latency, no vendor lock
•Git as transport layer is genuinely clever: orphan branches distribute team intent history with zero infrastructure

Weaknesses

•Beta stage, works only with Claude Code (adoption risk if LLM/agent landscape shifts)
•Niche audience: requires both AI-assisted workflows AND team git discipline to justify overhead

Post Description

Two problems with AI coding today:

1. Every session starts from zero. The agent sees your code but has no idea why it looks the way it does. 2. Team knowledge stays siloed. What one developer's agent learned yesterday doesn't help another's today. Rekal captures session context (turns, tool calls, file changes) at every commit into a local DuckDB database. A 2-10 MB session file compresses down to ~300 bytes on the wire via a custom binary codec with zstd and string interning. It shares across your team via git orphan branches — no extra infrastructure.

Search is three-way hybrid (BM25 + LSA + nomic-embed-text), all inside a single binary. No external APIs, no cloud, no setup. The embedding model ships in the binary.

The agent controls how much context it loads — progressive retrieval keeps token cost minimal.

Other design choices: append-only with content-hash dedup (merge conflicts are structurally impossible), ~200ms search at 14k turns, zero runtime dependencies beyond git. Written in Go, Apache-2.0.